“I just think it cannot go forward without there being broad satisfaction with the way the company has addressed money laundering...data protection, consumer privacy….” FRB Chairman Jerome H. Powell, Testimony to House Financial Services Committee, July 10, 2019.
“We have an open mind, but not an open door.” Bank of England Governor Mark Carney, June 18, 2019.
“A wider use of new types of crypto-assets for retail payment purposes would warrant close scrutiny by authorities to ensure that they are subject to high standards of regulation.” Letter to G20 Leaders, FSB Chair (and FRB Vice Chair) Randal K. Quarles, June 24, 2019.
Facebook’s June 18 announcement that it has created a Geneva-based entity with plans to issue a currency called Libra is sending shock waves through the financial world. Within days, the Group of Seven announced the creation of a study group. And key central bank regulators, like FRB Chairman Jerome Powell, have subsequently warned against a “sprint to implementation” (see quote above).
The stated objectives of creating Libra are to improve the efficiency of payments, reducing costs and speeding transfers; and to improve financial access. While these are laudable goals, it is essential that we achieve them without facilitating criminal exploitation of the payments system or reducing the ability of authorities to monitor and mitigate systemic risk. In addition, any broad-based financial innovation should facilitate the stabilization of consumption.
On all of these criteria, we see Libra as doing more harm than good. It does not simply take advantage of Facebook’s platform to improve payments efficiency. In fact, Libra appears to be Facebook’s third entry into the payments world, the first two (Credits and Messenger) having been wound down or scaled back (see here and here). Yet, the third time appears to be anything but a charm: Libra could facilitate various criminal uses of finance, as well as become a significant source of systemic risk. Not only that, but for those in advanced economies who use it, Libra will likely add uncertainty to the purchasing power of savings. And, for the countries whose currencies are excluded from the Libra portfolio, it will diminish seignorage, while enabling capital outflows and, in periods of stress, accelerating capital flight.
Like Governor Carney (see quote above), we have an open mind, and believe that increased competition, coupled with the introduction of new technologies, will eventually lower stubbornly high transactions costs, improving the quality of financial services globally. But in this case, we urge a closed door.
Libra is a call-to-action for the official community. Financial regulators and central banks must now act quickly and decisively to ensure that evolving financial technologies are a positive force in society, rather than a threat to basic values and financial stability. We hope that the Financial Stability Board (FSB) will follow through aggressively on the “close scrutiny” that Chairman Quarles espouses (see quote above).
Turning to a few specifics, the idea behind Libra is that every individual with a smart phone will be able to purchase tokens using their own domestic currency. The Libra Association (a Geneva-based company) will invest proceeds from the sales of the tokens in a portfolio of bonds and in bank deposits denominated in what we assume to be a few major currencies. Owners can then use their tokens to purchase goods and services, or to extinguish debts. Assuming that this occurs on the Facebook platform, with 2.4 billion users worldwide, it becomes a mechanism where something like half of the world’s adult population can transact directly.
Will Libra really make payments faster, cheaper and safer both domestically and internationally? As we discuss in an earlier post, until recently, a combination of monopoly power and regulatory barriers has made payments slow and expensive. The entry of specialized companies like Venmo and Square, and the response of incumbent banks—creating transfer systems like Zelle—are cutting costs and speeding small payments. As far as we can tell, within most advanced economies, the vast majority of the population already has access to low cost payments technologies (like bank debit cards). Will Libra be cheaper?
For cross-border remittances, which often serve the least well off in society, the costs remain extremely high. As we suggested in a post on the subject, the costs of guarding against criminal activity are probably a key factor. Less expensive systems like Transferwise, or even Paypal, transfer funds from one bank account to another, relying on the banks’ compliance operations to “know your customer” and ensure against money laundering and terrorist finance. That is, systems are cheap when someone else bears the high fixed costs of verifying that criminals are neither sending nor receiving the funds. Libra will surely require the same licenses and compliance systems as existing providers. (For example, Western Union reportedly has money services licenses in more than 30 countries and reporting obligations in more than 50.) Will Libra be cheaper?
Similarly, we strongly agree with the goal of improving financial access. At last count, there were 1.7 billion unbanked adults in the world (see here). Bringing them into the financial system, as well as improving services for the underbanked, will yield large economic and social benefits. In an earlier post, we highlighted the amazing progress in India, where more than 350 million people have obtained no-frills bank accounts over a period of less than five years. The creation of these accounts and their increasing usage reflects government efforts to lower the costs of knowing your customer (Aadhaar requires everyone 15 years or older to obtain a 12-digit identification number tied to their biometric data), to subsidize basic access, and to shift from cash to electronic transfers. Can Libra do better?
So far, so bad. It gets worse. Why would people living in a country whose policymakers have achieved price stability want to hold savings in the form of a basket of currencies when the bulk of their purchases are in their domestic currency? A key purpose of savings is to allow people to smooth their consumption. That, in turn, requires stable purchasing power, but the value of Libra will move with the value of the currencies in its basket. To give some sense of how big such fluctuations might be, consider the case of the SDR―the IMF’s reserve asset that is a weighted average of the U.S. dollar, euro, Japanese yen, British pound, and Chinese yuan (the last was added in 2016). The following chart displays the value of the SDR in U.S. dollars (black line), as well as the five-year moving standard deviation of its annualized daily percentage change (red line). Note first that, while the jurisdictions whose currencies are included in the SDR generally enjoyed relatively low inflation since the mid-1990s, its U.S. dollar value fluctuated in a wide range between $0.95 and $1.65. Furthermore, the standard deviation of the return is between roughly 4 and 6½ percentage points.
Value of the SDR in U.S. dollars (Daily with rolling five-year standard deviation), 1981-July 2019
Source: IMF and authors’ calculations.
Why should law-abiding households and firms accept this added volatility in the purchasing power of their savings? In our view, if they understand the risks, they would not do so. The incentive to use Libra would be far greater in economies where purchasing power is less stable. But, as we discuss below, we doubt that the governments in these jurisdictions would (or should) allow domestic use of Libra.
As an aside, we note that tax authorities are likely to get involved in this enterprise from the start. Even if, as Libra’s documentation indicates, the net interest income from the Libra reserve accrues to its investors, the holders of Libra will experience gains and losses from currency fluctuations (like those of the SDR). These will surely require record keeping, reporting and tax payments. For those who comply with these obligations, the associated accounting complexities will likely make the use of Libra especially onerous for small payments. And, once again, it is difficult to see why any government would wish to relax its tax rules to encourage substitution away from use of its own currency in its domestic payments system.
Perhaps most important, were it to be successful, Libra seems very likely to become a source of systemic risk for the global financial system. First, if Libra (on Facebook or WhatsApp) replaces a significant fraction of the existing payments system, then its failure would be catastrophic for those who rely on it. Second, if the Libra Association guarantees convertibility into a national currency, then the Libra Reserve will operate as an open-ended mutual fund (where the Association investors, not the Libra holders, receive the interest). The current proposal states that the fund’s assets will be liquid government securities and bank deposits, with the Association settling the details. Like any open-ended fund, it would be subject to runs on any less-liquid assets, such as bank deposits (see here). The Libra Association could have significant bargaining power in obtaining higher deposit rates (especially from banks dependent on short-term wholesale finance). The more the Libra reserve invests in bank deposits, the more a Libra run would become a global bank run. Were the reserve to accumulate $1 trillion in assets—roughly the average of one of the top 10 U.S. bank holding companies—a run would precipitate a fire sale with global consequences. We need fewer, not more, structures like this.
Finally, for three reasons, we believe that it would be unwise for any country whose currency is not in the Libra basket to allow its citizens to use the new currency. First, the shift from domestic currency to Libra would reduce the government’s seignorage revenue, damaging public finances. Second, if domestic residents are transacting in Libra, the central bank risks losing control of the country’s monetary system (and its payments system), making it difficult to maintain price stability. And third, acquiring Libra rather than a domestic asset constitutes a capital outflow, which seems likely to increase interest rates and reduce domestic investment. Not only that, but Libra could become a vehicle for capital flight and a currency attack if people were to lose confidence in a country’s government finances, or its financial system more generally.
We doubt anyone really knows where this is going. What we do know is that Libra is fundamentally different from cryptocurrencies like Bitcoin. First, its backing will have fundamental value. Second, as the designers of Libra clearly state, the technology does not yet exist to make Libra permissionless (like Bitcoin) and still be useful for speeding billions of payments: as we discuss here, permissionless systems are limited in what they can do. Third, a consortium of for-profit companies, not a haphazard group of independent miners, controls Libra. Absent strong regulation, what incentive will this group have to protect customer identities, prevent misuse by criminals, and internalize its impact on the financial system as a whole? Why should we trust Facebook and its collaborators with such critical tasks?
While we may not know what Libra will become, we do know that it is not a mechanism for merely taking advantage of Facebook’s enormous platform to facilitate transactions domestically and cross-border. If Facebook believed that its platform could speed and lower the costs of exchange, in line with the hopes for Big Tech (see the latest BIS Annual Economic Report), then it could (like nonbanks Square, Venmo and Transferwise) simply partner with traditional banks around the world. Such a straightforward approach would exploit the comparative advantage of the two businesses, bringing together the network externalities of Facebook’s platform and the banks’ high-fixed-cost compliance and regulatory systems. There would be no need to introduce a new currency for this purpose. The lack of an effective, permissionless technology would be no obstacle. And, there would be very little, if any, regulatory uncertainty.
That brings us back to regulation, and the need for action. Regulators will have to decide how to treat Libra and how to ensure that the approach is consistent across jurisdictions. Their choices will undoubtedly set precedent for future Big Tech entries in the payments space, so the goal must be to promote innovation that strengthens competition and lowers costs without expanding threats to the financial system or compromising basic values.
Based on its economic function, Libra ultimately is one of three things: a bank deposit, a mutual fund, or a form of e-money (like M-Pesa), none of which pays a yield. If it is a bank deposit, then the Libra Association needs a banking license and must abide by banking laws wherever it operates. This would mean everything from capital and liquidity regulation to anti-money laundering systems and resolution planning. Furthermore, it would have to create a banking subsidiary in every jurisdiction where it takes deposits. If Libra is a mutual fund, then it would need to comply with asset management regulations and licensing requirements in every jurisdiction where it is offered. If it were to become so large as to be systemic, it could still be designated a “financial market utility,” making it subject to strict scrutiny by the Federal Reserve. If it is an e-money, in each jurisdiction where it is used it will need the approval of both financial services regulators who monitor payments system compliance and the central bank that needs to ensure the effectiveness of domestic monetary control. (Even in Kenya, the home of M-Pesa, some policymakers now wish to compel mobile money firms to become banks.)
Libra poses a variety of other issues beyond the many we have already listed. For example, what would happen to central bank balance sheets and to money multipliers if Libra leads to withdrawals from the domestic banking system? What about its impact on sovereign bond markets? If Libra were to become a significant force in those markets, able to influence risk-free yields in various jurisdictions, how would government debt managers react? The list goes on.
Perhaps the ultimate challenge Libra poses comes from the fact that it aspires to operate across multiple jurisdictions without a true “home.” For the most part, payments systems are set up to handle domestic institutions operating within national boundaries. We regulate these institutions primarily using domestic laws, with voluntary cooperation across borders. Even “global” banks have a home supervisor that examines them on a consolidated basis. The prospect of Libra serves to dramatize the existing asymmetry between this largely domestic approach and the rise of global, multi-jurisdictional entities that could choose any home temporarily and could shift at will.
The bottom line: Facebook’s proposed creation of Libra is a wake-up call. It makes inescapable the need for rapid, international coordination of financial regulation. Without mechanisms to ensure consistent application of coherent global rules, we may find ourselves in a world full of “low-quality finance havens” that exist to evade financial regulations—just like “tax havens” that exist merely to evade taxes.
Acknowledgment: We thank our friend and Stern colleague, Hanna Halaburda, for her very helpful suggestions.